HEDGEFUNDIE's excellent adventure Part II: The next journey
-
- Posts: 65
- Joined: Fri Oct 18, 2019 10:08 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Why not all UPRO?
I'm sure this has been asked multiple times so I'll try to add substance (I've read a lot of this thread and either haven't seen or don't remember these points discussed in this way so please correct me if otherwise).
Assumptions:
1. Adventurers are setting aside a portion of their net worth that wouldn't meaningfully impact their life if it disappeared.
2. The time horizon is long enough that the market is very likely to go up overall during the investment period.
It is given that a flat market, especially a volatile one, is bad for a leveraged ETF, so clearly there are periods of an overall positive market where UPRO loses money. But it seems like long term UPRO will likely end up ahead of SPY.
The purpose of TMF/EDV is to mitigate drawdown, while gaining a nice rebalancing bonus. The rebalancing bonus seems peripheral, as we expect UPRO to do better, so rather than being the reason for inclusion, it's just a nice bonus. But TMF/EDV is also a drag to overall returns, both with its lower expected return, and the drag of rebalancing into it in a bull market. So the point is safety. But my issue, and this has been discussed before obviously, is what is the utility of a "safety" element that can itself be blown up? Since hyperinflation would destroy it, it would also take down a significant chunk of UPRO, entirely separate from a bad equity market, just in the rebalancing into a falling knife.
(I'm not saying anything new here so its more for a summary of how I'm thinking about this.)
The Adventure is a bet that hyperinflation won't reoccur, and that to me seems like a good bet. But situations where 100% UPRO wouldn't beat this also seem pretty unlikely. It seems harder to construct a long term scenario where UPRO would die than where TMF would. I'd be interested to see some simulated returns before the start date for the fund, such as 1999-2009.
I wonder, what if people considering this first thought of how much of their portfolio they'd be willing to devote to UPRO/TMF, then just cut that in half and put it into UPRO, basically whether UPRO/SPY wouldn't be a better bet than UPRO/TMF, with SPY representing the half not invested and remaining in one's normal equity instead.
It seems far more interesting to hold UPRO/TMF (or EDV) so I hope I'm wrong. But the idea of holding TMF as risk mitigation seems contradictory to the notion that this whole thing is meant to be a lottery ticket, the risk mitigation is the rest of the portfolio, so why not aim even higher with only UPRO and maybe or maybe not reduce the size of the overall allocation to compensate?
I'm sure this has been asked multiple times so I'll try to add substance (I've read a lot of this thread and either haven't seen or don't remember these points discussed in this way so please correct me if otherwise).
Assumptions:
1. Adventurers are setting aside a portion of their net worth that wouldn't meaningfully impact their life if it disappeared.
2. The time horizon is long enough that the market is very likely to go up overall during the investment period.
It is given that a flat market, especially a volatile one, is bad for a leveraged ETF, so clearly there are periods of an overall positive market where UPRO loses money. But it seems like long term UPRO will likely end up ahead of SPY.
The purpose of TMF/EDV is to mitigate drawdown, while gaining a nice rebalancing bonus. The rebalancing bonus seems peripheral, as we expect UPRO to do better, so rather than being the reason for inclusion, it's just a nice bonus. But TMF/EDV is also a drag to overall returns, both with its lower expected return, and the drag of rebalancing into it in a bull market. So the point is safety. But my issue, and this has been discussed before obviously, is what is the utility of a "safety" element that can itself be blown up? Since hyperinflation would destroy it, it would also take down a significant chunk of UPRO, entirely separate from a bad equity market, just in the rebalancing into a falling knife.
(I'm not saying anything new here so its more for a summary of how I'm thinking about this.)
The Adventure is a bet that hyperinflation won't reoccur, and that to me seems like a good bet. But situations where 100% UPRO wouldn't beat this also seem pretty unlikely. It seems harder to construct a long term scenario where UPRO would die than where TMF would. I'd be interested to see some simulated returns before the start date for the fund, such as 1999-2009.
I wonder, what if people considering this first thought of how much of their portfolio they'd be willing to devote to UPRO/TMF, then just cut that in half and put it into UPRO, basically whether UPRO/SPY wouldn't be a better bet than UPRO/TMF, with SPY representing the half not invested and remaining in one's normal equity instead.
It seems far more interesting to hold UPRO/TMF (or EDV) so I hope I'm wrong. But the idea of holding TMF as risk mitigation seems contradictory to the notion that this whole thing is meant to be a lottery ticket, the risk mitigation is the rest of the portfolio, so why not aim even higher with only UPRO and maybe or maybe not reduce the size of the overall allocation to compensate?
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
In case you missed it, the following is from the very first post of the original thread:RussellWilson wrote: ↑Thu Apr 01, 2021 10:44 pm Why not all UPRO?
I'm sure this has been asked multiple times so I'll try to add substance (I've read a lot of this thread and either haven't seen or don't remember these points discussed in this way so please correct me if otherwise).
And from a bit later on in the original thread:HEDGEFUNDIE wrote: ↑ Why not just 100% UPRO?
Because the drawdowns will be super deep during market crashes, and it may take decades to recover. Here is the backtest between 40/60 3xS&P/3xLTT (Portfolio 1) and 100% 3xS&P (Portfolio 2).
HEDGEFUNDIE wrote: ↑ Next, let's look at the new UPRO data compared to the S&P 500 going all the way back to 1955:
Of course, this ride would not have been for the faint-of-heart. Here is a comparison of the drawdowns:
-
- Posts: 65
- Joined: Fri Oct 18, 2019 10:08 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Thanks. I'd say 97.5% drawdown is probably not ideal. It's quite something seeing a real life example of (what would have been) UPRO pretty much disappearing.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Yes, I'm glad you see this quickly. I've seen other posters failing to recognize how crazy 97.5% DD is.RussellWilson wrote: ↑Thu Apr 01, 2021 11:24 pm Thanks. I'd say 97.5% drawdown is probably not ideal. It's quite something seeing a real life example of (what would have been) UPRO pretty much disappearing.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Especially when that draw down happens 10 or 15 years into the experience and HFEA comprises 80 or 90% of your Net Worth...RussellWilson wrote: ↑Thu Apr 01, 2021 11:24 pm Thanks. I'd say 97.5% drawdown is probably not ideal. It's quite something seeing a real life example of (what would have been) UPRO pretty much disappearing.
Better lucky than smart.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I think the risk of disappearing would be greater if the bulk of its AUM came from long-term holders like those on HFEA.RussellWilson wrote: ↑Thu Apr 01, 2021 11:24 pm Thanks. I'd say 97.5% drawdown is probably not ideal. It's quite something seeing a real life example of (what would have been) UPRO pretty much disappearing.
But instead, it's intended to be held for less than a day. Same reason why the volatility ETPs stick around reverse split after reverse split: so long as people continue to day-trade them, they will stay open.
-
- Posts: 37
- Joined: Sat Mar 06, 2021 4:07 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
just wanted to give an update on this -- i've been trying out the SMA (and sma + bands) strategy on quantconnect, which has minute/second data. if you're going to automate the strategy through quant, then this is probably a lot more accurate backtest than the daily data my app was pulling from yahoo. historical backtesting is much more difficult on that platform though, so i kinda thumbnail with leveraged older securities to see how they'd survive the 80s, 90s, 00s, 10s, etc. and then i try to the 10s on quantconnect.RovenSkyfall wrote: ↑Wed Mar 31, 2021 1:39 pm Yeah the backtesting isn't intended to correctly predict the future. It is just an attempt to show how a particular strategy performed in certain situations in the past and how those responses compare to the alternatives.
It seems like preparing for multiple likely potential futures is more valuable than preparing for just the best possible future.
I have no idea what TMF is going to do in the next 10 years. It clearly has room to go down. It has room to go up.
I do know that the declining rate environment lead to TMF significantly adding to the return of the HFEA (and makes that historical CAGR really attractive), but I am doubtful (but unsure) that TMF will continue to have that secondary effect. I agree with HEDGEFUNDIE that recent economic policies seem more likely to be repeated than the policies of the 70s, 80s or even 90s. The Fed seems to like to reduce rates in a downturn (at least for the most recent past ones), which makes me think TMF is still of value.
The nebulous long term expected returns of TMF make me wonder if there might be a better way to utilize it. The concept that interests me the most is the 200dMA of the SP500. It seems like a good signal of when it might be an opportune time to get out of UPRO and into something else like TMF. With a 5% band (or even without) the number of trades per year only slightly increases over a quarterly rebalanced HFEA.
The potential pitfall of being 100% UPRO when SMA is above 200d is that you could get caught when the tide goes out if you miss your signal. A mix of 80/20% UPRO/TMF retains some small benefit of TMF if the tide rapidly goes out. Additionally, one could transition to HFEA 55/45 when the SP500 drops below the 200dMA. Uncorrelated showed at the end of this post that binary on off can be worse for the CER than being in some combination of the two funds. The ultimate goal would be to hold less TMF in the good times when UPRO is driving returns while keeping the benefit of TMF if UPRO crashes.
a few interesting things i've found:
1) on my own app, y2k absolutely kills tech stocks, which makes upro much better than nasdaq with a 1980-2020 strategy. even the sma doesn't help (though i didn't try running a different signal for it).
2) there is a negligible difference between 2x and 3x bond leverage CAGR since 1980, with a significant increase in drawdown from 3x (e.g. people should consider using UBT instead of TMF) - the ticker i tried it with is PRULX
2) for minute-by-minute trades, i think 4% bands is best. but overfit at your own risk.
Last edited by millennialblues on Fri Apr 02, 2021 12:31 pm, edited 1 time in total.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I always thought that was just CYA language in the prospectus. Is the majority of the AUM really holding for less than a day?BayStater wrote: ↑Fri Apr 02, 2021 8:18 amI think the risk of disappearing would be greater if the bulk of its AUM came from long-term holders like those on HFEA.RussellWilson wrote: ↑Thu Apr 01, 2021 11:24 pm Thanks. I'd say 97.5% drawdown is probably not ideal. It's quite something seeing a real life example of (what would have been) UPRO pretty much disappearing.
But instead, it's intended to be held for less than a day. Same reason why the volatility ETPs stick around reverse split after reverse split: so long as people continue to day-trade them, they will stay open.
- RovenSkyfall
- Posts: 267
- Joined: Wed Apr 01, 2020 11:40 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I may have missed it, but did we ever find out if the borrowing rates for TMF went back down to a more reasonable rate?langlands wrote: ↑Mon Oct 05, 2020 11:48 amHi coingaroo, could you link the documents where you find the financing costs for TMF, UPRO, and TQQQ? You seem to think TMF (1.2%) is more expensive to borrow than UPRO and TQQQ (0.8%) whereas Uncorrelated and others seem to think it's the reverse.coingaroo wrote: ↑Wed Sep 23, 2020 11:16 am My biggest concern right now is TMF's yields combined with absurdly high financing costs. I am waiting for the annual report of TMF, coming around October 31st. I want to see just exactly how much interest they are paying on swaps.
The April report said that they were paying apparently 1.2% in interest to borrow for TMF; which is absolutely ludicrous when the LIBOR rate was 0.37% that day. Keep in mind that TMF borrows >250% of its value, only about 50% are in direct long term treasuries in $TLT (which are the collateral).
If you are in UPRO/TQQQ; you don't have to worry. Their borrowing costs were far more reasonable, at like 0.8% or so on the same day.
Furthermore, TMF's swaps are a leverage on the $TLT ETF fund; with a ER of 0.15%. I do not believe this is reported in the TMF SEC ER because they can technically say that it's not an acquired fund expense, they're just buying swaps on the TLT index and it just turns out the TLT index is deflated by TLT ERs.
Looking at the fund fact sheet, they say acquired fund expenses are 0.13%, which is clearly too low for 0.15% x 3 (due to the 3x leverage). So, it seems pretty clear that they are only counting the direct TLT holdings' acquired expenses; and the TLT ER deflator is an additional (undisclosed) drag on returns.
Anyways, let's look at all the fees you are paying with TMF:
Total real expense ratio of TMF = 4.425% + volatility drag
- 0.15% * 2.5 in the form of TLT ER tracking drag on swaps = 0.375%
1.2% * 2.5 in the form of swap financing premiums = 3%
1.05% in the form of TMF ER = 1.05%
I don't care about historical performance, when 30 year treasuries were paying 3.45% (Jan 2019 ), 10.35% - (4.425% + libor adjustment) - vol drag = positive.
But now with a yield to maturity of 1.34%, 4.02% - 4.425% - vol drag, it would be an absolute no go. It is an asset with a negative expected value, in nominal terms even, and crazy stdev. You never invest in things with negative expected values: that is called losing money.
I am just hoping the 1.2% swap financing costs (which you pay ~2.5x) was a fluke on that particular day (April 30th) the report was generated, and real financing costs have came down dramatically. Otherwise, I will be replacing TMF.
Thanks.
I saved my money, but it can't save me | The Chariot
-
- Posts: 6
- Joined: Sat Feb 20, 2021 9:13 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Probably trivial for most of you guys but I made a google spread sheet for my upro/tmf/vixy portfolio. Telling me when I should rebalance(according to bands) and how much. Feel free to copy it and let me know if something is wrong.
https://docs.google.com/spreadsheets/d/ ... p=drivesdk
https://docs.google.com/spreadsheets/d/ ... p=drivesdk
-
- Posts: 1594
- Joined: Mon Aug 24, 2020 5:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
http://direxioninvestments.onlineprospe ... ETF-AR.pdfRovenSkyfall wrote: ↑Fri Apr 02, 2021 3:38 pmI may have missed it, but did we ever find out if the borrowing rates for TMF went back down to a more reasonable rate?langlands wrote: ↑Mon Oct 05, 2020 11:48 amHi coingaroo, could you link the documents where you find the financing costs for TMF, UPRO, and TQQQ? You seem to think TMF (1.2%) is more expensive to borrow than UPRO and TQQQ (0.8%) whereas Uncorrelated and others seem to think it's the reverse.coingaroo wrote: ↑Wed Sep 23, 2020 11:16 am My biggest concern right now is TMF's yields combined with absurdly high financing costs. I am waiting for the annual report of TMF, coming around October 31st. I want to see just exactly how much interest they are paying on swaps.
The April report said that they were paying apparently 1.2% in interest to borrow for TMF; which is absolutely ludicrous when the LIBOR rate was 0.37% that day. Keep in mind that TMF borrows >250% of its value, only about 50% are in direct long term treasuries in $TLT (which are the collateral).
If you are in UPRO/TQQQ; you don't have to worry. Their borrowing costs were far more reasonable, at like 0.8% or so on the same day.
Furthermore, TMF's swaps are a leverage on the $TLT ETF fund; with a ER of 0.15%. I do not believe this is reported in the TMF SEC ER because they can technically say that it's not an acquired fund expense, they're just buying swaps on the TLT index and it just turns out the TLT index is deflated by TLT ERs.
Looking at the fund fact sheet, they say acquired fund expenses are 0.13%, which is clearly too low for 0.15% x 3 (due to the 3x leverage). So, it seems pretty clear that they are only counting the direct TLT holdings' acquired expenses; and the TLT ER deflator is an additional (undisclosed) drag on returns.
Anyways, let's look at all the fees you are paying with TMF:
Total real expense ratio of TMF = 4.425% + volatility drag
- 0.15% * 2.5 in the form of TLT ER tracking drag on swaps = 0.375%
1.2% * 2.5 in the form of swap financing premiums = 3%
1.05% in the form of TMF ER = 1.05%
I don't care about historical performance, when 30 year treasuries were paying 3.45% (Jan 2019 ), 10.35% - (4.425% + libor adjustment) - vol drag = positive.
But now with a yield to maturity of 1.34%, 4.02% - 4.425% - vol drag, it would be an absolute no go. It is an asset with a negative expected value, in nominal terms even, and crazy stdev. You never invest in things with negative expected values: that is called losing money.
I am just hoping the 1.2% swap financing costs (which you pay ~2.5x) was a fluke on that particular day (April 30th) the report was generated, and real financing costs have came down dramatically. Otherwise, I will be replacing TMF.
Thanks.
Page 117.
0.4 - 0.5% borrowing cost
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Is that APY? WOWTingting1013 wrote: ↑Sat Apr 03, 2021 6:51 pmhttp://direxioninvestments.onlineprospe ... ETF-AR.pdfRovenSkyfall wrote: ↑Fri Apr 02, 2021 3:38 pmI may have missed it, but did we ever find out if the borrowing rates for TMF went back down to a more reasonable rate?langlands wrote: ↑Mon Oct 05, 2020 11:48 amHi coingaroo, could you link the documents where you find the financing costs for TMF, UPRO, and TQQQ? You seem to think TMF (1.2%) is more expensive to borrow than UPRO and TQQQ (0.8%) whereas Uncorrelated and others seem to think it's the reverse.coingaroo wrote: ↑Wed Sep 23, 2020 11:16 am My biggest concern right now is TMF's yields combined with absurdly high financing costs. I am waiting for the annual report of TMF, coming around October 31st. I want to see just exactly how much interest they are paying on swaps.
The April report said that they were paying apparently 1.2% in interest to borrow for TMF; which is absolutely ludicrous when the LIBOR rate was 0.37% that day. Keep in mind that TMF borrows >250% of its value, only about 50% are in direct long term treasuries in $TLT (which are the collateral).
If you are in UPRO/TQQQ; you don't have to worry. Their borrowing costs were far more reasonable, at like 0.8% or so on the same day.
Furthermore, TMF's swaps are a leverage on the $TLT ETF fund; with a ER of 0.15%. I do not believe this is reported in the TMF SEC ER because they can technically say that it's not an acquired fund expense, they're just buying swaps on the TLT index and it just turns out the TLT index is deflated by TLT ERs.
Looking at the fund fact sheet, they say acquired fund expenses are 0.13%, which is clearly too low for 0.15% x 3 (due to the 3x leverage). So, it seems pretty clear that they are only counting the direct TLT holdings' acquired expenses; and the TLT ER deflator is an additional (undisclosed) drag on returns.
Anyways, let's look at all the fees you are paying with TMF:
Total real expense ratio of TMF = 4.425% + volatility drag
- 0.15% * 2.5 in the form of TLT ER tracking drag on swaps = 0.375%
1.2% * 2.5 in the form of swap financing premiums = 3%
1.05% in the form of TMF ER = 1.05%
I don't care about historical performance, when 30 year treasuries were paying 3.45% (Jan 2019 ), 10.35% - (4.425% + libor adjustment) - vol drag = positive.
But now with a yield to maturity of 1.34%, 4.02% - 4.425% - vol drag, it would be an absolute no go. It is an asset with a negative expected value, in nominal terms even, and crazy stdev. You never invest in things with negative expected values: that is called losing money.
I am just hoping the 1.2% swap financing costs (which you pay ~2.5x) was a fluke on that particular day (April 30th) the report was generated, and real financing costs have came down dramatically. Otherwise, I will be replacing TMF.
Thanks.
Page 117.
0.4 - 0.5% borrowing cost
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Well, talk about a coincidence; I was just thinking, a few days ago, that I should do a mini Excel spreadsheet to make my life easier when rebalancing time comes around. Not trivial at all, and much welcome! Thank you!Purple_Light wrote: ↑Sat Apr 03, 2021 6:31 pm Probably trivial for most of you guys but I made a google spread sheet for my upro/tmf/vixy portfolio. Telling me when I should rebalance(according to bands) and how much. Feel free to copy it and let me know if something is wrong.
https://docs.google.com/spreadsheets/d/ ... p=drivesdk
-
- Posts: 1594
- Joined: Mon Aug 24, 2020 5:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
1 month LIBOR end October 2020 was 0.15% so TMF borrrows through Total Return Swaps at between 30 and 40 bp over LIBOR. Those are annual rates obviously
Better lucky than smart.
-
- Posts: 6
- Joined: Sat Feb 20, 2021 9:13 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
BienvenueOohLaLa wrote: ↑Sat Apr 03, 2021 10:34 pmWell, talk about a coincidence; I was just thinking, a few days ago, that I should do a mini Excel spreadsheet to make my life easier when rebalancing time comes around. Not trivial at all, and much welcome! Thank you!Purple_Light wrote: ↑Sat Apr 03, 2021 6:31 pm Probably trivial for most of you guys but I made a google spread sheet for my upro/tmf/vixy portfolio. Telling me when I should rebalance(according to bands) and how much. Feel free to copy it and let me know if something is wrong.
https://docs.google.com/spreadsheets/d/ ... p=drivesdk
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I haven’t read the full thread. What is the most standard implementation of this strategy?
Is it 55% UPRO / 45% TMF with quarterly rebalancing? Or what is the most common rebalancing strategy?
Is it 55% UPRO / 45% TMF with quarterly rebalancing? Or what is the most common rebalancing strategy?
- firebirdparts
- Posts: 4414
- Joined: Thu Jun 13, 2019 4:21 pm
- Location: Southern Appalachia
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Thanks
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
And this is where I regret doing this in a taxable account. I’m sitting 63/37 and didn’t do my rebalance on the 1st. I don’t really want to sell some UPRO but also don’t want to add more funds to the strategy.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
YTD - finally in the green
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
How much Roth space do you have? Assuming your capital gains are still relatively modest, it might be worth it to nip the problem in the bud and reimplement in a tax-exempt account. Tax-deferred could also be an option and solves the rebalancing problem. But in the good timeline where the strategy performs very well, you might end up with a bloated IRA or 401K that's difficult to withdraw from tax-efficiently. In any case, it's probably a good idea to figure out what the right thing to do is now before the problem gets bigger.bjcleaver wrote: ↑Mon Apr 05, 2021 2:53 pmAnd this is where I regret doing this in a taxable account. I’m sitting 63/37 and didn’t do my rebalance on the 1st. I don’t really want to sell some UPRO but also don’t want to add more funds to the strategy.
-
- Posts: 1594
- Joined: Mon Aug 24, 2020 5:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
-
- Posts: 386
- Joined: Sun Feb 04, 2018 6:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
My opinion:Tingting1013 wrote: ↑Mon Apr 05, 2021 3:56 pmExcept that if you had been in cash instead of TMF for those 2 years, you would be much worse off.
- It is a fallacy that bonds are safe. There is a reason why the 60/40 portfolios don't beat the SP500 - it's the bonds!!
- TMF should only be viewed as insurance against UPRO in the event that there is a prolonged slow bear market with falling or stagnant interest rates.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
You aren't wrong, but the question is what do you do to hedge UPRO when not covered by TMF.perfectuncertainty wrote: ↑Mon Apr 05, 2021 5:55 pm My opinion:
- It is a fallacy that bonds are safe. There is a reason why the 60/40 portfolios don't beat the SP500 - it's the bonds!!
- TMF should only be viewed as insurance against UPRO in the event that there is a prolonged slow bear market with falling or stagnant interest rates.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Don't disagree and look at this: 50/50 UPRO/TMF PVperfectuncertainty wrote: ↑Mon Apr 05, 2021 5:55 pm My opinion:
- It is a fallacy that bonds are safe. There is a reason why the 60/40 portfolios don't beat the SP500 - it's the bonds!!
- TMF should only be viewed as insurance against UPRO in the event that there is a prolonged slow bear market with falling or stagnant interest rates.
Click on exposures and look at Risk Contribution on the far right. If I am interpreting this correctly, TMF is a lot risker then UPRO is. UPRO has 79% of the return, but only 47.75% of the portfolio risk. TMF has 21% of the return and 52.25% of the portfolio risk.
(I understand the insurance aspect of it, but this is really bad insurance for the risk you take on - please correct me if I am reading the Risk Contribution wrong).
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I can correct you. The timeframe you have is the longest bull market in history, so UPRO looks rosier than the reality.ljford7 wrote: ↑Mon Apr 05, 2021 6:21 pmDon't disagree and look at this: 50/50 UPRO/TMF PVperfectuncertainty wrote: ↑Mon Apr 05, 2021 5:55 pm My opinion:
- It is a fallacy that bonds are safe. There is a reason why the 60/40 portfolios don't beat the SP500 - it's the bonds!!
- TMF should only be viewed as insurance against UPRO in the event that there is a prolonged slow bear market with falling or stagnant interest rates.
Click on exposures and look at Risk Contribution on the far right. If I am interpreting this correctly, TMF is a lot risker then UPRO is. UPRO has 79% of the return, but only 47.75% of the portfolio risk. TMF has 21% of the return and 52.25% of the portfolio risk.
(I understand the insurance aspect of it, but this is really bad insurance for the risk you take on - please correct me if I am reading the Risk Contribution wrong).
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Maybe you can help with this then. If you plug in UBT (ProShares Ultra 20+ Year Treasury 2X) in lieu of TMF, your return % of both UBT and UPRO aligns almost perfectly with risk contributions.Marseille07 wrote: ↑Mon Apr 05, 2021 6:23 pm I can correct you. The timeframe you have is the longest bull market in history, so UPRO looks rosier than the reality.
TMF/UPRO for the same time period still shows the large gap in return/risk. PV Link
Shouldn't UBT has the same issue as TMF in the return versus risk due to the bull market?
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
In another thread a math error came to light in the calculation of simulated leveraged ETF returns.
Financial decisions based on emotion often turn out to be bad decisions.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I haven't looked at UBT closely because the ER is 0.95%, which I find too high. I'm kind of picky in that I don't look at much of something I don't intend to trade.ljford7 wrote: ↑Mon Apr 05, 2021 6:32 pmMaybe you can help with this then. If you plug in UBT (ProShares Ultra 20+ Year Treasury 2X) in lieu of TMF, your return % of both UBT and UPRO aligns almost perfectly with risk contributions.Marseille07 wrote: ↑Mon Apr 05, 2021 6:23 pm I can correct you. The timeframe you have is the longest bull market in history, so UPRO looks rosier than the reality.
TMF/UPRO for the same time period still shows the large gap in return/risk. PV Link
Shouldn't UBT has the same issue as TMF in the return versus risk due to the bull market?
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?ljford7 wrote: ↑Mon Apr 05, 2021 6:21 pmDon't disagree and look at this: 50/50 UPRO/TMF PVperfectuncertainty wrote: ↑Mon Apr 05, 2021 5:55 pm My opinion:
- It is a fallacy that bonds are safe. There is a reason why the 60/40 portfolios don't beat the SP500 - it's the bonds!!
- TMF should only be viewed as insurance against UPRO in the event that there is a prolonged slow bear market with falling or stagnant interest rates.
Click on exposures and look at Risk Contribution on the far right. If I am interpreting this correctly, TMF is a lot risker then UPRO is. UPRO has 79% of the return, but only 47.75% of the portfolio risk. TMF has 21% of the return and 52.25% of the portfolio risk.
(I understand the insurance aspect of it, but this is really bad insurance for the risk you take on - please correct me if I am reading the Risk Contribution wrong).
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
by tradri » Fri Apr 02, 2021 3:59 pm
chris319 wrote: ↑Fri Apr 02, 2021 3:34 pm
pct_change = (pct_change - expense_ratio / 252) * leverage
Now you've got me confused. Should the exp ratio be subtracted before or after leverage is applied? Your code vs. mine. In the first instance (yours) it seems like the exp ratio is being multiplied by the leverage.
capital = capital * ((((pl(ct1) - 1) * 100 * leverage / 100) + 1) - expRatio)
You're right. The expense ratio shouldn't be multiplied by the leverage. It should only be subtracted once from the daily leveraged return divided by 255.
Guess I was using a wrong simulation then...that sucks.
Financial decisions based on emotion often turn out to be bad decisions.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
"So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
Better lucky than smart.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Honestly, that's the way I think we should view TMF. It's crash insurance protection, it may provide some return to the portfolio or it may be a drag but it should be there when crash happens. Though, now the question is, is there better crash insurance or at least cheaper ones? That open up a whole new can of worms regarding Vix and maybe alternatively method of holding (some sort of trend following??? momentum investing???).taojaxx wrote: ↑Mon Apr 05, 2021 7:00 pm "So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Not much Roth space but trying to rectify that with some non-deductible conversions from my IRA. It’ll just take a while to build up a balance.langlands wrote: ↑Mon Apr 05, 2021 3:55 pm
How much Roth space do you have? Assuming your capital gains are still relatively modest, it might be worth it to nip the problem in the bud and reimplement in a tax-exempt account. Tax-deferred could also be an option and solves the rebalancing problem. But in the good timeline where the strategy performs very well, you might end up with a bloated IRA or 401K that's difficult to withdraw from tax-efficiently. In any case, it's probably a good idea to figure out what the right thing to do is now before the problem gets bigger.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
After working extensively on a dynamic version of VIX hedging (to avoid being eaten up by contango), I dropped the idea due to complexity and the constant attention required.jarjarM wrote: ↑Mon Apr 05, 2021 7:05 pmHonestly, that's the way I think we should view TMF. It's crash insurance protection, it may provide some return to the portfolio or it may be a drag but it should be there when crash happens. Though, now the question is, is there better crash insurance or at least cheaper ones? That open up a whole new can of worms regarding Vix and maybe alternatively method of holding (some sort of trend following??? momentum investing???).taojaxx wrote: ↑Mon Apr 05, 2021 7:00 pm "So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
I settled for a reduced TMF allocation (30%) and a rotational momentum based UPRO as defined here:
viewtopic.php?f=10&t=297591
Better lucky than smart.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I read that thread as well. While I'm not totally on board with the 200 day SMA, I can see the appeal, especially since TMF expected upside is essentially none.taojaxx wrote: ↑Mon Apr 05, 2021 7:28 pmAfter working extensively on a dynamic version of VIX hedging (to avoid being eaten up by contango), I dropped the idea due to complexity and the constant attention required.jarjarM wrote: ↑Mon Apr 05, 2021 7:05 pmHonestly, that's the way I think we should view TMF. It's crash insurance protection, it may provide some return to the portfolio or it may be a drag but it should be there when crash happens. Though, now the question is, is there better crash insurance or at least cheaper ones? That open up a whole new can of worms regarding Vix and maybe alternatively method of holding (some sort of trend following??? momentum investing???).taojaxx wrote: ↑Mon Apr 05, 2021 7:00 pm "So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
I settled for a reduced TMF allocation (30%) and a rotational momentum based UPRO as defined here:
viewtopic.php?f=10&t=297591
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
We never know though, TMF could still surprise us lol:jarjarM wrote: ↑Mon Apr 05, 2021 7:38 pmI read that thread as well. While I'm not totally on board with the 200 day SMA, I can see the appeal, especially since TMF expected upside is essentially none.taojaxx wrote: ↑Mon Apr 05, 2021 7:28 pmAfter working extensively on a dynamic version of VIX hedging (to avoid being eaten up by contango), I dropped the idea due to complexity and the constant attention required.jarjarM wrote: ↑Mon Apr 05, 2021 7:05 pmHonestly, that's the way I think we should view TMF. It's crash insurance protection, it may provide some return to the portfolio or it may be a drag but it should be there when crash happens. Though, now the question is, is there better crash insurance or at least cheaper ones? That open up a whole new can of worms regarding Vix and maybe alternatively method of holding (some sort of trend following??? momentum investing???).taojaxx wrote: ↑Mon Apr 05, 2021 7:00 pm "So what's the takeaway here? That in this portfolio UPRO is actually insurance for TMF?
Top
TMF is crash insurance: it is only useful when stocks crater and the Fed cuts aggressively, sending bond prices spiking. The rest of the time, as all insurance policies, it costs money. The past 40 years it has been an insurance policy adding to returns, but that's an anomaly.
I settled for a reduced TMF allocation (30%) and a rotational momentum based UPRO as defined here:
viewtopic.php?f=10&t=297591
viewtopic.php?f=10&t=278700
Better lucky than smart.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
On to negative rate!!!taojaxx wrote: ↑Mon Apr 05, 2021 8:13 pm
We never know though, TMF could still surprise us lol:
viewtopic.php?f=10&t=278700
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Even if one would... I doubt these or similar etfs will exist throughout 30 years continuously.
Get rich or die tryin'
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
DCA-ing into a mix of TMF, EV and VXX has been pretty painful on this end for a few months.Ramjet wrote: ↑Wed Mar 31, 2021 2:46 pmI agree with thiscorp_sharecropper wrote: ↑Wed Mar 31, 2021 2:31 pmMentally it seems "easier", to me, to take a certain amount of money and let it ride as it's simple to always think of it as just that original sum even during a terrifying drawdown. Once you start continuously DCAing and have no idea what you've been putting in, I imagine it's easier to start thinking of a lot of the growth as "your" money which makes draw downs more painful.
Get rich or die tryin'
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I have an off-the-wall question.
I've been looking into how much predictive information is contained in the market history. The idea is to see if there are reasonable methods for selecting rotations among funds.
Volatility is known to cluster. I decided to look at trailing volatility to see if that gives any information on returns; for example, 21-day average volatility as a measure for predicting 21-day leading returns.
I see that there is a pretty strong relationship between trailing volatility and the spread in future returns. I would expect that behavior if volatility clusters.
I was expecting to see negative (or at least reduced) returns with an increase in trailing volatility, on the theory that increased volatility is a symptom of fear before and during market drops. Instead, I tend to see trends towards increased mean and median returns with increased volatility for funds like UPRO and TQQQ. I get the behavior of decreased returns with increased volatility for TMF. For smaller sectors, the trend is much flatter.
Has anybody seen such behavior? It seems counter intuitive, maybe I'm doing something wrong.
I've been looking into how much predictive information is contained in the market history. The idea is to see if there are reasonable methods for selecting rotations among funds.
Volatility is known to cluster. I decided to look at trailing volatility to see if that gives any information on returns; for example, 21-day average volatility as a measure for predicting 21-day leading returns.
I see that there is a pretty strong relationship between trailing volatility and the spread in future returns. I would expect that behavior if volatility clusters.
I was expecting to see negative (or at least reduced) returns with an increase in trailing volatility, on the theory that increased volatility is a symptom of fear before and during market drops. Instead, I tend to see trends towards increased mean and median returns with increased volatility for funds like UPRO and TQQQ. I get the behavior of decreased returns with increased volatility for TMF. For smaller sectors, the trend is much flatter.
Has anybody seen such behavior? It seems counter intuitive, maybe I'm doing something wrong.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
That does seem somewhat counterintuitive. Are you looking at data from last 10 years or longer term? We do know that some of the highest return dates are also cluster with some of the worst return days (see March 2020) so it be interesting to see if this due to dataset issue or actual behavior issue.Hydromod wrote: ↑Mon Apr 05, 2021 10:00 pm I have an off-the-wall question.
I've been looking into how much predictive information is contained in the market history. The idea is to see if there are reasonable methods for selecting rotations among funds.
Volatility is known to cluster. I decided to look at trailing volatility to see if that gives any information on returns; for example, 21-day average volatility as a measure for predicting 21-day leading returns.
I see that there is a pretty strong relationship between trailing volatility and the spread in future returns. I would expect that behavior if volatility clusters.
I was expecting to see negative (or at least reduced) returns with an increase in trailing volatility, on the theory that increased volatility is a symptom of fear before and during market drops. Instead, I tend to see trends towards increased mean and median returns with increased volatility for funds like UPRO and TQQQ. I get the behavior of decreased returns with increased volatility for TMF. For smaller sectors, the trend is much flatter.
Has anybody seen such behavior? It seems counter intuitive, maybe I'm doing something wrong.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I’m splicing together the longest series I can. UPRO is extended with VFINX, TQQQ with QQQ and NDX, and so on. The series that go back only to the early 2000s are noisier but still seem to have a similar behavior.
I use every day in the sequence as a separate observation, on the theory that each event might be encountered at a different offset in practice.
I use every day in the sequence as a separate observation, on the theory that each event might be encountered at a different offset in practice.
-
- Posts: 386
- Joined: Sun Feb 04, 2018 6:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
IMO holding TMF constantly might be the incorrect approach.
Take a look at a 3-year Sharp Chart of SPY and TLT.
Try this: whenever TLT is below SPY then do not own TMF or scale it way back. When TLT is above SPY, then move to a 50/50 position of UPRO and TMF.
For reference, Sharp Charts can be found here on Stockcharts
Take a look at a 3-year Sharp Chart of SPY and TLT.
Try this: whenever TLT is below SPY then do not own TMF or scale it way back. When TLT is above SPY, then move to a 50/50 position of UPRO and TMF.
For reference, Sharp Charts can be found here on Stockcharts
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Maybe the leveraged character skews the volatility=fear relationship and makes it less reliable. That's one reason UPRO rotational strategies use SPY as a signal and not UPRO itself.Hydromod wrote: ↑Mon Apr 05, 2021 10:00 pm I have an off-the-wall question.
I've been looking into how much predictive information is contained in the market history. The idea is to see if there are reasonable methods for selecting rotations among funds.
Volatility is known to cluster. I decided to look at trailing volatility to see if that gives any information on returns; for example, 21-day average volatility as a measure for predicting 21-day leading returns.
I see that there is a pretty strong relationship between trailing volatility and the spread in future returns. I would expect that behavior if volatility clusters.
I was expecting to see negative (or at least reduced) returns with an increase in trailing volatility, on the theory that increased volatility is a symptom of fear before and during market drops. Instead, I tend to see trends towards increased mean and median returns with increased volatility for funds like UPRO and TQQQ. I get the behavior of decreased returns with increased volatility for TMF. For smaller sectors, the trend is much flatter.
Has anybody seen such behavior? It seems counter intuitive, maybe I'm doing something wrong.
Better lucky than smart.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Problem with this is it's an "ex post" rule: a number of times SPY was "above" TLT right before the accident happened and you would own no TMF heading into the crash. That's why I keep a 30% TMF allocation at all times despite following a 200 DMA UPRO rotational allocation.perfectuncertainty wrote: ↑Mon Apr 05, 2021 10:20 pm IMO holding TMF constantly might be the incorrect approach.
Take a look at a 3-year Sharp Chart of SPY and TLT.
Try this: whenever TLT is below SPY then do not own TMF or scale it way back. When TLT is above SPY, then move to a 50/50 position of UPRO and TMF.
For reference, Sharp Charts can be found here on Stockcharts
March 2020 was a good example: UPRO (and SPY) flying high heading into the wall.
Better lucky than smart.